OGSA Teleconference - 25 July 2005 ================================== * Participants Mike Behrens (R2AD, LLC) Hiro Kishimoto (Fujitsu) Tom Maguire (IBM) Takuya Mori (NEC) Andreas Savva (Fujitsu) Dave Snelling (Fujitsu) Jem Treadwell (HP) Jay Unger (IBM) Pete Ziu (Northrop Grumman) Apologies: Fred Maciel, Steven Newhouse Minutes: Andreas Savva * Minutes of July 20 approved with no changes * August F2F update First version of agenda sent out by Hiro: https://forge.gridforum.org/projects/ogsa-wg/document/OGSA-2005-Aug-F2F-agenda/en/1 Venue: Fujitsu Laboratories America, Sunnyvale, CA - The meeting will be held in the same building as last year, but in a different, larger, room. - An additional room is reserved for Tuesday morning for the OGSA-EGA meeting (parallel session). - Internet access will be available. - There are more participants expected overall but many people will probably attend some sessions only. - The agenda is still in an early draft stage. Session leaders are expected to provide an agenda for their sessions. - OGSA-AuthZ WG participants have not been contacted yet. - Jem does not know if he can attend yet. * Teleconference schedule review Hiro put together a proposal for the teleconference schedule based on next items discussion (previous call) https://forge.gridforum.org/projects/ogsa-wg/document/telecon_schedule/en/26 - Provisioning: There is some ongoing work within other bodies (e.g., DMTF Utility Computing WG, EGA). OGSA-WG should work to make clear how this work fits with other efforts. (Any overlapping effort at this moment is probably 'people overlap'. That is, the same people may be involved in more than one group.) - Aug 10: The plan is to invite OGSA-AuthZ to join the teleconference. This is in part to make sure the F2F is productive. - Agreed to move the OGSA 1.5 discussion from Wed (7/27) to Monday (8/3) to give Andreas and Jem some more time to prepare. - Agreed to move the Roadmap from 8/27 to 8/10. It is not expected that a revision of the Roadmap document will take a long time. - There is time allocated to resource modelling starting next week. * WSRF Basic Profile 1.0 review Based on version 30 uploaded by Tom before the call: https://forge.gridforum.org/projects/ogsa-wg/document/draft-ggf-ogsa-wsrf-basic-profile/en/25 Tom deleted existing comments from the document and put them in trackers. [Issue numbers are tracker artifact numbers on the gridforge BP tracker: https://forge.gridforum.org/tracker/index.php?func=browse&group_id=42&atid=780] - Issue 1571: TLS extensions not in WS-I BSP 1.0 Hiro has confirmed that they are present in the BSP 1.0. Closed. - Issue 1559: l.483 remaining issue is the combination of 'only...permitted'. Agreed that for consistency with the rest of the text it should be changed to 'only....recommended.' Fixed. - Sec.6.1.1 Updated R0611 and R0612 to use the terms defined in the referenced WS-N documents: UseNotify changed to UseRaw; Filter, etc. R0617, R0618: deleted by accident. Rejected deletion and made similar terminology changes to bring up to date with current WS-N specification. The R06XX statements need re-numbering for consistency. - l.630: URI reference was deleted since it was not cited in the text. - Some URIs in the document were hyperlinks. Removed the hyperlinks. - SchemaCentricCanonicalization issues - Issues raised on serialization by Takuya in the last call. Some were retracted after a re-read of the canonicalization specification. (In particular the inefficient serialization issue.) - The remaining issue is with defining an envelope for the query response to hold the type of the XPath return value. - The issuer of the query should know what the return value is going to be. Anyway this information is best described in the data not the message. - Agreed that serialization is essential for interoperability and has to be defined in the BP. But that this extra definition is nice-to-have and probably best left to higher levels (JAX-RPC etc.) - GGF IPR in the schema - Consensus that it is ok to leave the full version in. It is somewhat verbose but it is on the safe side. It can be changed later if GGF comes up with a guideline. - R0423: the namespace should be wsrf-rpw - Issue 1242: verified and closed - Issue 1250: verified and closed - Issue 1251: verified and closed - Issue 1255: verified and closed - Issue 1256: verified and closed - Issue 1258: R0431: minor inconsistency: Change MUST to SHOULD - Verify and close in the next version - Issue 1294: Verified and closed. - But note that this spec refers to WS-I BSP 1.0 dated 20-Jan-2005. WS-I had updated the date of this spec but has not changed the version number (new date is 11-May-2005). Agreed to leave the citation as is. Action: Takuya will check for differences between the two versions and will report back to the group. - Issue 1320: verified and closed - Issue 1355: verified and closed - Issue 1360: Closed (as obsolete; insufficient description.) Note that the Base Faults PR spec has an extensibility element in the schema that causes indeterminism. A later draft has fixed it but it cannot be referenced yet. The extensibility point should be addded to Sec.7.1. Also to add a restriction that only type extension is allowed. And that no extensibility elements are allowed. Action: Tom has the pen and will update the document with the above changes. Action: ALL should check FIXED trackers before the next call to expedite the verification process. * Next call - WSRF BP final call - Review changes - Tracker review and verification.